Unemployment Rates Around the World 2020

The COVID-19 pandemic will eventually go away; the consequences for jobs and livelihoods across the globe will be felt for many years to come. This is the main takeaway from the latest World Economic Outlook released by the International Monetary Fund in October. The pandemic had—and will continue to have—especially severe effects on the most economically vulnerable people, informally employed women and younger workers in particular. The burden of the crisis, the report says, has fallen unevenly across economic sectors: while the impact of the recession has been less grievous for those able to work from home, workers employed in industries such as accommodation and food services, transportation, retail and wholesale have been especially hard hit. Furthermore, the pandemic exacerbated pre-existing trends in poverty and income inequality which was already on the rise in many advanced, emerging and developing economies. Lockdowns and school closures cast a long shadow on millions of children’s future prospects too: a trove of academic studies has demonstrated that lower and interrupted lifetime schooling is associated with lower lifetime income and earnings trajectories. The toll of COVID-19, quite simply, threatens to fully undo decades of progress and push tens of millions of people into job insecurity, not just today but tomorrow as well.

The scenario depicted by the economists at the Fund, it is important to highlight, stands in stark contrast with that of the global financial crisis of 2007-2009, when the top quintile of the income scale (meaning the upper-middle class) bore the brunt of a recession triggered by the housing speculative bubble in the US and the cascade of banking and corporate failures that followed. While not all recessions are the same, they all tend to result in rapidly rising unemployment rates that take a very long time to fall after positive economic growth returns.

But what does unemployment mean? It is a question that appears easy to answer, at least superficially. Not being able to afford rent, to get an education or visit a doctor, to care for yourself and your family: unemployment, we know, has many ramifications. However, translating each individual situation into data, and data into policies that can improve the situation of millions of individuals, is remarkably arduous. While experts agree that the jobless rate represents the percentage share of the labor force out of work, and that high unemployment ratios can threaten growth and social cohesion, they often disagree when it comes to measuring joblessness. There are multiple ways to appraise the job market’s myriad realities.

The official unemployment rate is determined by calculated by dividing the number of unemployed individuals by all individuals in the labor force. The trouble starts when it comes to figuring out who exactly is and is not in the labor force? The very individuals in question often cannot tell whether they should consider themselves employed or unemployed.

For example: a person who loses a well-compensated full-time job and settles for a part-time position that pays a fraction is by default classified as “employed,” while another person who actively seeks work but takes a few weeks off from the search is not even counted as part of the labor force. An individual who would like to work but is unable due to a disability or medical condition is in the very same position.

The result is that many economists believe that—because of the existence of persons who are unemployed or hidden under-employed—statistics are inherently skewed and paint a too-rosy picture (yes, even during these gloomy times) of the workforce. Needless to say, unemployment and hidden underemployed too are very difficult to measure.

Tracking the labor market is made even more complicated when different tracking tools tell different stories. Whether through census-type methods, employment office records, surveys of a sample of the population or multi-approach techniques, their conclusions will only offer an approximate reflection of the economic and social health of a country.

Nevertheless, over time, unemployment rates remain a crucial indicator of the health, level of development and growth trajectory of an economy. Rising unemployment results in loss of income for individuals and reduced collection of taxes for governments, forcing them to spend greater amounts on unemployment benefits and social subsidies. Long-term unemployment can also weaken the strength of the social fabric, lead to mass frustration and rejection of democratic political orders, prompt cross-border migrations or threaten the economy of trading partners.

In conclusion then—taking into account the many reservations about the accuracy of workforce tracking methods—how many jobs have been lost globally since the beginning of the COVID-19 pandemic? The IMF tries to answer this question by referencing estimates by the UN’s International Labour Organization (ILO). Assuming a 48-hour working week, the reduction in work hours in the second quarter of 2020 was equivalent to the loss of 495 million full-time jobs, which added to the equivalent of 160 million full-time jobs lost in the first quarter. The ILO expects the losses to pile up as the year nears to the end, with a decline in working hours in the third quarter equal to 345 million full-time positions—19.8% of which in the Americas, 12.4% in the Arab states, 11.6% in Europe and Central Asia, followed by Africa at 11.5% and the Asia-Pacific region at 10.7%. Looking further ahead, in a baseline scenario, an estimated equivalent of 245 million full-time jobs could be lost in the fourth quarter (whereas the pessimistic view projects a loss corresponding to 515 million posts, and the “optimistic” view to 160 million). Gone in the blink of an eye, many of these work hours will take months or years to come back while otheres might not even exist anymore once the pandemic is behind us.

Whereas such broad numbers give us a hint of where jobs are (or no longer are) today, they often suggest little about their nature and where they will go eventually. Experts argue that in the span of just a few months, COVID-19 rapidly accelerated developments that were slowly becoming mainstream—the increase in remote working, the digitization of many processes and the replacement of full-time employees with contingent workers being the most obvious ones. The pandemic has also renewed fears that automation will replace entire job categories: robots can assemble car parts, robots can scrub floors, and robots can pick up vegetables. 

In pre-pandemic research, the McKinsey Global Institute studied more than 2,000 work activities focusing on 46 countries representing about 80% of the global workforce and quantified the technical feasibility of automating each of them. The proportion of occupations that can be fully automated using demonstrated technology, McKinsey concluded, is actually small: less than 5%. However, as it became clear over these past several months, even if whole occupations are not automated, partial automation is set to affect almost all occupations to a greater or lesser degree, with about 60% of them having at least 30% of activities that can be performed by machines. In a more recent survey of company executives around the world, McKinsey confirmed that, due to the pandemic, the adoption of automation has accelerated “moderately” or “significantly” in almost 7 businesses out of 10 among those examined.

So, should we resign ourselves to a future of high unemployment and widening social inequality? The truth, as a famous quote goes, is that prediction is always very difficult, especially if it’s about the future. The example of those handful of countries that have managed to bring the number of new COVID-19 cases to near zero and their economic activity and unemployment rates to near pre-pandemic levels, should guide us and give us hope. Jobs are lost, jobs are created, jobs are learned. Hang tight.

*Values are expressed in terms of a percentage.

Source: International Monetary Fund, World Economic Outlook Database, October 2020.